AIDeepSeek-AI and Peking University open-source DSpark, using speculative decoding to boost LLM inference throughput by up to 400%The method also accelerates Gemma and Qwen model inference.digg it—5 Sources, 123d ago, first seen 123d agoCombined views37.2K5 Sources, first seen 123d ago407 likesCombined views37.2K5 Sources, first seen 123d ago407 likes
